When a project finishes late and the parties disagree about why, the schedule becomes evidence. Each side commissions analysis, each analysis produces an answer, and the answers rarely agree. Much of that disagreement can be traced to a single early decision: which method of delay analysis to use.
Choosing a method is not a technical formality. Different methods ask different questions of the same record, and they are sensitive to different weaknesses in it. Choose well and the analysis is clear, reproducible and persuasive. Choose poorly and even accurate work can be dismissed.
The family of methods
Recognized guidance, including AACE International’s recommended practice on forensic schedule analysis and the Society of Construction Law Delay and Disruption Protocol, describes a range of approaches. They differ mainly in whether they look forward or back, and in which schedules they rely on.
- As-planned versus as-built: comparing the original plan with what actually happened
- Impacted as-planned: adding delay events to the original plan to estimate their effect
- Time impact analysis: inserting delay events into the schedule update current at the time
- Windows and contemporaneous period analysis: tracing the critical path period by period through the updates
- Collapsed as-built: removing delay events from the as-built record to identify a but-for completion
Each has a sound theoretical basis. Each also has conditions under which it performs badly.
Let the record decide
The single most important factor in choosing a method is the quality of the project record. A method that depends on reliable contemporaneous updates is only as good as those updates. If they were poorly maintained, manipulated or simply missing, a method that leans on them will inherit every flaw.
- Are the schedule updates complete, regular and internally consistent?
- Do daily reports, correspondence and meeting minutes corroborate the dates?
- Is the analysis prospective, supporting a time extension request, or retrospective, explaining what happened?
- What does the contract require for demonstrating entitlement?
- Where will the analysis be tested: in negotiation, mediation, arbitration or court?
Time impact analysis is often specified for prospective requests during the project. Windows analysis is powerful retrospectively when the updates can be trusted. Collapsed as-built can be the most reliable choice when updates are weak but the as-built record is strong. The method should follow the facts.
The strongest delay analysis is the one the other side can reproduce and still cannot fault.
Concurrency is where analyses are won or lost
Concurrent delay, where owner-caused and contractor-caused delays affect completion over the same period, is the most contested ground in most disputes. A method that cannot identify concurrency clearly invites challenge. The analyst must be able to show, period by period, which delays were driving completion and whether they overlapped, and then apply the contract and governing law to the result.
This is also where judgment matters most. Software can identify paths. It cannot decide whether a delay was truly critical or merely coincident. That determination needs an analyst who understands both the work and the rules being applied.
What makes an analysis survive scrutiny
Analyses that hold up under cross-examination share a recognizable profile. They are transparent, with every assumption stated and every adjustment to the schedule documented. They are grounded in contemporaneous evidence, not reconstruction. They identify the critical path from the logic rather than by assertion. And they acknowledge their own limitations, which makes the conclusions more credible, not less.
Analyses that fail tend to share a profile too: methods chosen to produce a desired answer, undisclosed changes to logic, and confidence that outruns the evidence.
Prevention is the better method
The best protection against a difficult delay dispute is a project that keeps its record well from the start. A sound baseline, regular and honest updates, timely notices and a simple delay event log do more for a future claim than any amount of retrospective analysis. They also make disputes less likely, because when both sides can see the same picture, there is less to argue about.
A practical starting point
If a dispute is likely, bring in delay expertise early, ideally while the project is still running. An early review of the record can identify the method that will best fit it, close gaps while witnesses and documents are still available, and give counsel a realistic view of the strength of the position. The cost is modest. The value, when the question finally has to be answered, can be decisive.
